Gheorghe Tecuci

Seven Aspects of Mixed-Initiative Reasoning:An Introduction to this Special Issue on Mixed-Initiative Assistants

Mixed-initiative assistants are agents that interact seamlessly with humans to extend their problem-solving capabilities or provide new capabilities. Developing such agents requires the synergistic integration of many areas of AI, including knowledge representation, problem solving and planning, knowledge acquisition and learning, multiagent systems, discourse theory, and human-computer interaction. This paper introduces seven aspects of mixed-initiative reasoning (task, control, awareness, communication, personalization, architecture, and evaluation) and discusses them in the context of several state-of-the-art mixed-initiative assistants. The goal is to provide a framework for understanding and comparing existing mixed-initiative assistants and for developing general design principles and methods.

Automating knowledge acquisition as extending, updating, and improving a knowledge base

A method for the automation of knowledge acquisition that is viewed as a process of incremental extension, updating, and improvement of an incomplete and possibly partially incorrect knowledge base of an expert system is presented. The knowledge base is an approximate representation of objects and inference processes in the expertise domain. Its gradual development is guided by the general goal of improving this representation to consistently integrate new input information received from the human expert. The knowledge acquisition method is presented as part of a methodology for the automation of the entire process of building expert systems, and is implemented in the system NeoDISCIPLE. The method promotes several general ideas for the automation of knowledge acquisition, such as understanding-based knowledge extension, knowledge acquisition through multistrategy learning, consistency-driven concept formation and refinement, closed-loop learning, and synergistic cooperation between a human expert and a learning system. >

Apprenticeship learning in imperfect domain theories

This chapter presents DISCIPLE, a multistrategy, integrated learning system illustrating a theory and a methodology for learning expert knowledge in the context of an imperfect domain theory. DISCIPLE integrates a learning system and an empty expert system, both using the same knowledge base. It is initially provided with an imperfect (nonhomogeneous) domain theory and learns problem-solving rules from the problem-solving steps received from its expert user, during interactive problem-solving sessions. In this way, DISCIPLE evolves from a helpful assistant in problem solving to a genuine expert. The problem-solving method of DISCIPLE combines problem reduction, problem solving by constraints, and problem solving by analogy. The learning method of DISCIPLE depends on its knowledge about the problem-solving step (the example) from which it learns. In the context of a complete theory about the example, DISCIPLE uses explanation-based learning to improve its performance. In the context of a weak theory about the example, it synergistically combines explanation-based learning, learning by analogy, empirical learning, and learning by questioning the user, developing its competence. In the context of an incomplete theory about the example, DISCIPLE learns by combining the above-mentioned methods, improving both its competence and performance.

Learning based on conceptual distance

An approach to concept learning from examples and concept learning by observation is presented that is based on a intuitive notion of conceptual distance between examples (concepts) and combines symbolical and numerical methods. The approach is based on the observation that very different examples generalize to an expression that is very far from each of them, while identical examples generalize to themselves. Following this idea the authors propose some domain-independent and intuitively justified estimates for the conceptual distance. A hierarchical conceptual clustering algorithm that groups objects so as to maximize the cohesiveness (a reciprocal of the conceptual distance) of the clusters is presented. It is shown that conceptual clustering can improve learning from complex examples describing objects and the relation between them. >

Training and Using DISCIPLE Agents A Case Study in the Military Center of Gravity Analysis Domain

This article presents the results of a multifaceted research and development effort that synergistically integrates AI research with military strategy research and practical deployment of agents into education. It describes recent advances in the DISCIPLE approach to agent development by subject-matter experts with limited assistance from knowledge engineers, the innovative application of DISCIPLE to the development of agents for the strategic center of gravity analysis, and the deployment and evaluation of these agents in several courses at the U.S. Army War College.

THE DISCIPLE–RKF LEARNING AND REASONING AGENT

Over the years we have developed the Disciple theory, methodology, and family of tools for building knowledge‐based agents. This approach consists of developing an agent shell that can be taught directly by a subject matter expert in a way that resembles how the expert would teach a human apprentice when solving problems in cooperation. This paper presents the most recent version of the Disciple approach and its implementation in the Disciple–RKF (rapid knowledge formation) system. Disciple–RKF is based on mixed‐initiative problem solving, where the expert solves the more creative parts of the problem and the agent solves the more routine ones, integrated teaching and learning, where the agent helps the expert to teach it, by asking relevant questions, and the expert helps the agent to learn, by providing examples, hints, and explanations, and multistrategy learning, where the agent integrates multiple learning strategies, such as learning from examples, learning from explanations, and learning by analogy, to learn from the expert how to solve problems. Disciple–RKF has been applied to build learning and reasoning agents for military center of gravity analysis, which are used in several courses at the US Army War College.

An experiment in agent teaching by subject matter experts

This paper presents a successful knowledge-acquisition experiment in which subject matter experts who did not have any prior knowledge-engineering experience succeeded in teaching the Disciple- COA agent how to critique courses of action, a challenge problem addressed by the DARPAs High-Performance Knowledge Bases program. We first present the COA challenge problem and the architecture of Disciple- COA, a learning agent shell from the Disciple family. Then we present the knowledge acquisition experiment, detailing both the expert?Disciple interactions, and the automatic knowledge-base development processes that take place as a result of these interactions. The results of this experiment provide strong evidence that the Disciple approach is a viable solution to the knowledge acquisition bottleneck.

Plausible Justification Trees: A Framework for Deep and Dynamic Integration of Learning Strategies

This article describes a framework for the deep and dynamic integration of learning strategies. The framework is based on the idea that each single-strategy learning method is ultimately the result of certain elementary inferences (like deduction, analogy, abduction, generalization, specialization, abstraction, concretion, etc.). Consequently, instead of integrating learning strategies at a macro level, we propose to integrate the different inference types that generate individual learning strategies. The article presents a concept-learning and theory-revision method that was developed in this framework. It allows the system to learn from one or from several (positive and/or negative) examples, and to both generalize and specialize its knowledge base. The method integrates deeply and dynamically different learning strategies, depending on the relationship between the input information and the knowledge base. It also behaves as a single-strategy learning method whenever the applicability conditions of such a method are satisfied.

Building An Intelligent Business Process Reengineering System: A Case-Based Approach

The Intelligent Business Process Reengineering System (IBPRS) is an interactive expert system. This paper presents a framework for IBPRS which utilizes case-based planning and problem-solving techniques in identifying requirements and problems and in searching for alternative opportunities from previous experiences (cases). IBPRS comprises two major components: Planner and Constructor. The IBPRS Planners problem solving consists of four stages: problem formulation, case retrieval, case refinement, and evaluation. In generating business process reengineering (BPR) alternatives, IBPRS identifies appropriate case(s) among previous BPR cases that are represented in the form of process model(s) using search algorithms and other rules stored in the Planners knowledge base. In evaluating the generated BPR alternatives, various simulation models and economic analyses are used. The best BPR alternative is then selected through a multi-attribute criteria decision analysis. In the refinement stage, IBPRS employs case-based substitution method for case adaptation and refinement. The Planner Knowledge Base and the Case Repository are updated as new knowledge develops and new cases are built and entered into the system. The main benefits of IBPRS is to facilitate BPR efforts by identifying problems, evaluating alternatives, and choosing the most appropriate BPR alternative.

Teaching intelligent agents: the disciple approach

The ability to build intelligent agents is significantly constrained by the knowledge acquisition effort required. Many iterations by human experts and knowledge engineers are currently necessary to develop knowledge‐based agents with acceptable performance. We have developed a novel approach, called Disciple, for building intelligent agents that relies on an interactive tutoring paradigm, rather than the traditional knowledge engineering paradigm. In the Disciple approach, an expert teaches an agent through five basic types of interactions. Such rich interaction is rare among machine learning (ML) systems, but is necessary to develop more powerful systems. These interactions, from the point of view of the expert, include specifying knowledge to the agent, giving the agent a concrete problem and its solution that the agent is to learn a general rule for, validating analogical problems and solutions proposed by the agent, explaining to the agent reasons for the validation, and being guided to provide new k...